Product Definitions for Smart Manufacturing

Summary

To develop and deploy advances in standards, conformance testing, user-awareness, and adoption of 3D model-based product definition standards to improve product quality and reduce costs for manufacturers throughout the product lifecycle.

This project has concluded.  Current related research is in the Digital Thread for Manufacturing project.

Description

The Product Definitions for Smart Manufacturing project will deliver methods, protocols, and tools for developing, conformance testing, increasing user-awareness, and industrial adoption of product definition standards necessary for the digital transformation of manufacturing enterprises. Smart manufacturing research at NIST has promoted a vision of fully-integrated, collaborative manufacturing systems that respond in real time to meet changing demands and conditions in the factory, in the supply network, and in customer needs. This vision is increasingly achievable by small-to-medium sized enterprises due to development of increasingly capable standards for conveying industrial data. NIST has had success in developing and promulgating product definition standards for manufacturing, however, there is still a great need for NIST leadership and research to continue those efforts.

WHAT IS THE TECHNICAL IDEA?
Recent NIST pilot projects and proof-of-concept studies demonstrated that deployment of emerging model-based product definition standards reduces design-to-manufacturing cycle time and improves final part quality. Further, much progress has been made in the past 5 years in advancing a suite of product-definition standards (i.e., STEP, QIF and MTConnect) that enable the model-based enterprise. However, there remain capability gaps in the standards; barriers slowing standards development, including dwindling expertise; a lack of public documentation, reference data and reference software implementations that enable understanding, implementation and adoption of the standards. This project will provide research advances in product-definition standardization, conformance testing, and traceability of data assets.
Furthermore, OEM supply chains are global and not all the standards have the international buy-in required by global markets. All these needs will be considered as NIST provides technical leadership and guidance to move standards-development organizations to agile standards development methods, to promote understanding and adoption of standards, to harvest QIF into ISO and further harmonize its use with STEP standards. The stakeholder community will be expanded to include automotive manufacturers. JT, 3D PDF, and other visualization formats will be analyzed, enhanced, conformance tested within the context of a model-based enterprise.
Conformance testing of manufacturing information against relevant standards remains a challenge due to time consuming manual processes for verification and validation.  EXPRESS-based STEP instance files can be automatically checked for syntax and structure conformance and XML-based QIF files can be checked that they are well-formed. However, information encoded semantically in those files is documented only in text- and diagram-based documents. The semantic information in instance files also needs to be checked for conformance. Encoding text- and diagram-based rules is a manual process that requires deciphering their meaning and writing appropriate software code for conformance checking. Methodologies are needed to transform text- and diagram-based standards documentation to computer-processable forms.
Lastly, proper curation of digital assets is critical to daily operations and requires permanent access and maintenance of trustworthy data. Corruption of that data can have catastrophic consequences on product development and affect viability of an enterprise. There is a need to protect the product data and its owner(s) by providing authorization, authentication, and traceability of trustworthy product data through the product lifecycle. Currently, knowing who can use data, how the data can be used, and who did what to the data is mainly captured in contracts and manual paper-based tracking methods. Industry needs a faster, more secure and sustainable way to record, embed or link authentication, authorization, and traceability information to the product data. 

WHAT IS THE RESEARCH PLAN?
The research plan includes three areas of activity: 
Prior work from the Smart Manufacturing Operations Planning and Control Program’s Digital Thread for Smart Manufacturing Project will be continued. NIST will continue its leadership roles in PDES, Inc., ISO TC184 / SC 4, the Dimensional Metrology Standards Consortium, and increase participation in the 3D PDF Consortium. The current set of stakeholders will be expanded to include automotive manufacturers.  There are many tasks to accomplish this goal that relate to user requirements for standards, providing technical leadership in standards organizations and implementor forums, and publishing standards. 

• Identify product definition requirements of automotive manufacturers
• Identify gaps in critical standards and prioritize those gaps in standards roadmapping activities
• Develop agile and model-based methods to document and deploy standards
• Promote use of standards through better public-facing documentation, test data, reference software implementations
• Deliver STEP geometry in 3D PDF
• Publish ISO 10303-242 edition 2 FDIS and IS
• Publish Smart Manufacturing Reference Model specification through ISO/IEC JWG21 that meets US requirements
• Provide technical leadership to harvest QIF as an ISO specification

Research related to conformance testing will focus on semantic representations of product and manufacturing information (PMI) in product definitions standards for smart manufacturing and how they are tested.  PMI is used to communicate the allowable tolerances to manufacturing and inspection systems.  Research will develop new methodologies and software that seeks to provide automation to capture the intent of PMI defined in standards and the use of PMI in instance files.  

• Develop methodologies to process text- and diagram-based standards documentation with the goal of automatically generating code for checking instance files for conformance to the standard. Intermediate steps to achieve that goal are: investigating natural language processing, structured technical English, standard representations of rules, and translation of old diagram-based representations to modern information-based representations. 
• Develop methodologies for semantic conformance checking of instance files across many product data standards that share common information elements (such as PMI).
• Develop enhancements to the STEP File Analyzer and QIF File Analyzer based on user requirements, implementor forum input, and new features of the standards.
• Participate in implementor forums to test software implementations of STEP and 3D PDF.  

Traceability of product data with embedded authentication and authorization data is critical to smart manufacturing operations. There is a need to simplify and secure traceability through the product lifecycle, especially in complex supply chains of products with extensive lifespan. This task will:

• Evaluate blockchain, a distributed and replicated digital ledger that relies on proven cryptographic mechanisms and distributed consensus to control data quality. Blockchain is also virtually tamper resistant and has a native synchronization-discrepancy-resistance mechanism.
• Develop guidelines to implement a blockchain-based public and trustworthy tamper-resistant repository/registry, designed to enable and/or simplify product data traceability,
• Build a blockchain-based public and trustworthy tampering-resistant repository/registry to record product data transactions. 

Identifying valid transactions will help identify tampered data before it is used. Such a repository will: 
1) simplify the traceability of product data transactions due to the immutability of the records, 2) facilitate and automate pre-manufacturing fraud prevention to reduce the complex and expensive post-manufacturing faults detection, 3) reduce the number of faulty parts distributed, preventing brand reputation damage, loss of revenue from product returns, and reduce product liability issues.